Utilizing Artificial Intelligence in Software Testing
Mäkelä, Mikael (2019)
2019
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2019120123754
https://urn.fi/URN:NBN:fi:amk-2019120123754
Tiivistelmä
Artificial Intelligence has become more and more important part of computer science and IT business. Anyhow, it has not been used in the software testing widely.
At the same time the demand for using the test automation and testing more efficiently has increased. Using the AI and machine learning could be one approach to reduce the manual work in the software testing and also in the traditional test automation which still requires tremendous amunt of manual work. However, Artificial Intelligence and machine learning technologies still are pretty unknown for the major part of the software testing community.
The primary objective of the study was to form recommendations how the artificial intelligence could be utilized in the software testing. The secondary objective was to present the methods as part of standardized test process.
The research was conducted by creating the research keywords based on the testing standards and best practices. The different machine learning solutions related to software testing and its sub categories were searched both from the academic publications and online materials of the test tool providers, conference recordings and trainings provided by the AI testing specialists.
The actual search results are collected and categorized in four sections. In the first section as a theoretical background the AI and machine learning high level approaches have been presented and also challenges on testing the AI and with an AI.
In the second section the AI Testing model by AI for Software Testing Association is presented and reflected to a Paul Gerrard´s New Model for Testing.
In the third section the fundamental testing process has been used as a frame in which the different machine learning approaches from the other researches have been collected.
In the last section the different testing tools are presented and how the machine learning could be used to improve the efficiency using them. The section also presents some existing commercial and open source tools that are having macine learning features in their implementations.
At the same time the demand for using the test automation and testing more efficiently has increased. Using the AI and machine learning could be one approach to reduce the manual work in the software testing and also in the traditional test automation which still requires tremendous amunt of manual work. However, Artificial Intelligence and machine learning technologies still are pretty unknown for the major part of the software testing community.
The primary objective of the study was to form recommendations how the artificial intelligence could be utilized in the software testing. The secondary objective was to present the methods as part of standardized test process.
The research was conducted by creating the research keywords based on the testing standards and best practices. The different machine learning solutions related to software testing and its sub categories were searched both from the academic publications and online materials of the test tool providers, conference recordings and trainings provided by the AI testing specialists.
The actual search results are collected and categorized in four sections. In the first section as a theoretical background the AI and machine learning high level approaches have been presented and also challenges on testing the AI and with an AI.
In the second section the AI Testing model by AI for Software Testing Association is presented and reflected to a Paul Gerrard´s New Model for Testing.
In the third section the fundamental testing process has been used as a frame in which the different machine learning approaches from the other researches have been collected.
In the last section the different testing tools are presented and how the machine learning could be used to improve the efficiency using them. The section also presents some existing commercial and open source tools that are having macine learning features in their implementations.