Author
Şahin Gebizli, C., Metin, D., Sözer, Hasan
Publication Date
2015
Publication Place
-
HEART
Subject
Design verification, Software testing and verification, Test automation, Usage model based testing, Test efficiency, Usage profile, Test scenario creation, Code analysis, Software reliability
Type
Document
Language
Turkish
Digital
Yes
Manuscript
No
Library
Özyeğin University
Library Asset ID
1613-0073
Record ID
4bad9320-68b5-4daf-93e9-bfc7cf0a8661
Library Location
Computer Science
Date
2015
Sample Text
Model-based testing techniques increase efficiency by automatically creating test scenarios from the system usage model. In principle, it is possible to create an infinite number of test cases; however, resources to test these scenarios are limited.
Therefore, the content of the model used and the test case creation techniques should enable errors to be detected effectively. In this study, we propose a unique approach to improve the model content and model parameters used for model-based testing. The Markov chains we use in our approach are model based on statistical data.
It allows us to update the parameters to focus on scenarios with high error risk. Static code analysis techniques and usage
By evaluating profile analyses, we identify frequently used functions that are likely to encounter errors. We create the model content to test these functions. According to the dynamic analysis results, we update the model parameters to increase the probability of including functions that are prone to errors in the test scenarios created. When we used test scenarios created for a real Smart TV system software with this method, we observed that the error detection efficiency increased.
Cilt
1483
