Test generation can have a large impact on the software engineering process by decreasing the amount of time and effort required to maintain a high level of test coverage. This increases the quality of the resultant software while decreasing the associated effort. In this paper, we present TestNMT, an experimental approach to test generation using neural machine translation. TestNMT aims to learn to translate from functions to tests, allowing a developer to generate an approximate test for a given function, which can then be adapted to produce the final desired test.
We also present a preliminary quantitative and qualitative evaluation of TestNMT in both cross-project and within-project scenarios. This evaluation shows that TestNMT is potentially useful in the within-project scenario, where it achieves a maximum BLEU score of 21.2, a maximum ROUGE-L score of 38.67, and is shown to be capable of generating approximate tests that are easy to adapt to working tests.
Sun 4 NovDisplayed time zone: Guadalajara, Mexico City, Monterrey change
10:30 - 12:00 | |||
10:30 15mTalk | Total Recall, Language Processing, and Software Engineering NL4SE | ||
10:45 15mTalk | Is “Naturalness” a Result of Deliberate Choice? NL4SE Kevin Lee University of California at Davis, USA, Casey Casalnuovo University of California at Davis, USA | ||
11:00 15mTalk | A Fine-Grained Approach for Automated Conversion of JUnit Assertions to English NL4SE Danielle Gonzalez Rochester Institute of Technology, USA, Suzanne Prentice University of South Carolina, USA, Mehdi Mirakhorli Rochester Institute of Technology | ||
11:15 15mTalk | TestNMT: Function-to-Test Neural Machine Translation NL4SE | ||
11:30 15mTalk | 3CAP: Categorizing the Cognitive Capabilities of Alzheimer’s Patients in a Smart Home Environment NL4SE | ||
11:45 15mTalk | Generating Comments from Source Code with CCGs NL4SE | ||