Institutional Repository of Vilnius University
Savaiminio atsistatymo mechanizmo projektavimas ir efektyvumo vertinimas automatizuotų testų srityje /
Abstract
dc:descriptionThe thesis aims to design and empirically evaluate a self‑healing mechanism for Selenium‑ based UI tests, reducing maintenance overhead and false failures. A comparative study of existing tools (Healenium, Parasoft Selenic, Applitools, Testim, Mabl) revealed that most require project‑ specific tuning or produce many false positives in highly dynamic DOMs. To address this gap, a prototype named Healable WebDriver was implemented in Java 17 / Selenium 4. It applies a weighted locator search formula (ID, class, text, depth) and keeps the top‑K=3 candidates in a local JSON repository. Controlled experiments on five mutated Wikipedia DOM versions achieved 92.3 % accuracy, a 196 ms median healing time, and a 0.57 FP/FN ratio—statistically outperforming Healenium and Applitools (p < 0.05). These results confirm the hypothesis: the mechanism cuts locator‑fix time by at least 9 % and halves false alarms, improving CI stability. Future work should integrate semantic text similarity, auto‑tune weights, and extend the approach to mobile and API testing domains.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Deriugin, Vladimir,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD239030169&prefLang=en_US