Near Failure Analysis Using Dynamic Behavioural Data
Taromirad, Masoumeh LU and Runeson, Per LU
(2022)
23rd International Conference on Product-Focused Software Process Improvement, PROFES 2022
In Lecture Notes in Computer Science
13709.
p.171-178
- Abstract
Automated testing is a safeguard against software regression and provides huge benefits. However, it is yet a challenging subject. Among others, there is a risk that the test cases are too specific, thus making them inefficient. There are many forms of undesirable behaviour that are compatible with a typical program’s specification, that however, harm users. An efficient test should provide most possible information in relation to the resources spent. This paper introduces near failure analysis which complements testing activities by analysing dynamic behavioural metrics (e.g., execution time) in addition to explicit output values. The approach employs machine learning (ML) for classifying the behaviour of a program as faulty or healthy... (More)
Automated testing is a safeguard against software regression and provides huge benefits. However, it is yet a challenging subject. Among others, there is a risk that the test cases are too specific, thus making them inefficient. There are many forms of undesirable behaviour that are compatible with a typical program’s specification, that however, harm users. An efficient test should provide most possible information in relation to the resources spent. This paper introduces near failure analysis which complements testing activities by analysing dynamic behavioural metrics (e.g., execution time) in addition to explicit output values. The approach employs machine learning (ML) for classifying the behaviour of a program as faulty or healthy based on dynamic data gathered throughout its executions over time. An ML-based model is designed and trained to predict whether or not an arbitrary version of a program is at risk of failure. The very preliminary evaluation demonstrates promising results for feasibility and effectiveness of near failure analysis.
(Less)
- author
- Taromirad, Masoumeh
LU
and Runeson, Per
LU
- organization
- publishing date
- 2022-11-22
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Dynamic metrics, Failure prediction, Regression testing
- host publication
- Product-Focused Software Process Improvement : 23rd International Conference, PROFES 2022, Jyväskylä, Finland, November 21–23, 2022, Proceedings - 23rd International Conference, PROFES 2022, Jyväskylä, Finland, November 21–23, 2022, Proceedings
- series title
- Lecture Notes in Computer Science
- editor
- Taibi, Davide ; Kuhrmann, Marco ; Mikkonen, Tommi ; Abrahamsson, Pekka and Klünder, Jil
- volume
- 13709
- pages
- 8 pages
- publisher
- Springer Science and Business Media B.V.
- conference name
- 23rd International Conference on Product-Focused Software Process Improvement, PROFES 2022
- conference location
- Jyväskylä, Finland
- conference dates
- 2022-11-21 - 2022-11-23
- external identifiers
-
- scopus:85142706904
- ISSN
- 1611-3349
- 0302-9743
- ISBN
- 978-3-031-21388-5
- 978-3-031-21387-8
- DOI
- 10.1007/978-3-031-21388-5_12
- project
- Software Regression Testing with Near Failure Assertions
- language
- English
- LU publication?
- yes
- id
- 2f34923f-520e-41b6-a856-171cc5158fe1
- date added to LUP
- 2022-10-16 20:06:27
- date last changed
- 2024-05-30 20:52:46
@inproceedings{2f34923f-520e-41b6-a856-171cc5158fe1,
abstract = {{<p>Automated testing is a safeguard against software regression and provides huge benefits. However, it is yet a challenging subject. Among others, there is a risk that the test cases are too specific, thus making them inefficient. There are many forms of undesirable behaviour that are compatible with a typical program’s specification, that however, harm users. An efficient test should provide most possible information in relation to the resources spent. This paper introduces near failure analysis which complements testing activities by analysing dynamic behavioural metrics (e.g., execution time) in addition to explicit output values. The approach employs machine learning (ML) for classifying the behaviour of a program as faulty or healthy based on dynamic data gathered throughout its executions over time. An ML-based model is designed and trained to predict whether or not an arbitrary version of a program is at risk of failure. The very preliminary evaluation demonstrates promising results for feasibility and effectiveness of near failure analysis.</p>}},
author = {{Taromirad, Masoumeh and Runeson, Per}},
booktitle = {{Product-Focused Software Process Improvement : 23rd International Conference, PROFES 2022, Jyväskylä, Finland, November 21–23, 2022, Proceedings}},
editor = {{Taibi, Davide and Kuhrmann, Marco and Mikkonen, Tommi and Abrahamsson, Pekka and Klünder, Jil}},
isbn = {{978-3-031-21388-5}},
issn = {{1611-3349}},
keywords = {{Dynamic metrics; Failure prediction; Regression testing}},
language = {{eng}},
month = {{11}},
pages = {{171--178}},
publisher = {{Springer Science and Business Media B.V.}},
series = {{Lecture Notes in Computer Science}},
title = {{Near Failure Analysis Using Dynamic Behavioural Data}},
url = {{http://dx.doi.org/10.1007/978-3-031-21388-5_12}},
doi = {{10.1007/978-3-031-21388-5_12}},
volume = {{13709}},
year = {{2022}},
}