'Self-Healing Tests' - The holy grail of test automation?
summary
Claims that machine learning can reduce locator-driven failures by adapting tests at runtime using an open source stack: Healenium (Java/Selenium client + Docker backend + Postgres), an IntelliJ plugin and REST UI. Research cites 73.5% failing tests caused by bad locators; industry trends note ~40% ML adoption and ~90% rise in automation roles. A Digibank login POC shows Healenium producing a healed locator score ~79, re-running the find and turning the test green while recording heal metadata and reports.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 0:22SelfHealing Tests
- 1:28Who is Matthias
- 3:29Motivation
- 4:54Research
- 7:48The problem
- 9:32The solution
- 10:57Helium
- 13:49Helium components
- 16:10Example
- 20:15Report
- 23:47Questions
related talks
54mSelf-Healing Automation with Healenium and Regression Minimization with Drill4J
34mFuture of AI ML in Software Testing
25mSelf-Healing in Test Automation
54mStability in Unstability: Self-Healing Test Automation
55mStability in Unstability: Self‑Healing Test Automation
42mEnhancing Appium Tests' Stability with the Power of Healenium
31mSelf-Healing or Self-Killing?
32mSelf-healing system for UI tests using ML