Thesis - A Time Decaying Confidence Score for Micro-Testing

Computer Science; Software Testing and Quality Assurance, Software Engineering, Test Automation

Bachelor Thesis 22,5 hp or Master Thesis 30 hp

2 students completing 22,5 credits each or 2 master student completing 30 credits each.

Modern cloud-based software systems are composed of multiple services, APIs, and supporting infrastructure. In such systems, a change to one service can unintentionally affect user-facing functionality implemented across several components. Automated unit and integration tests can detect many defects, but they do not always provide a clear, feature-level view of whether a complete user workflow remains reliable.

This project investigates an approach for improving regression detection in a multi-service software system. The approach decomposes selected user-facing workflows into smaller actions, such as creating, updating, importing, or removing records. Each action is tested independently, and the results are combined into confidence scores for sub-features and then complete features.

The investigation is motivated by the potential to reduce reliance on repetitive manual acceptance testing while preserving confidence that important workflows continue to function correctly. However, the approach also introduces challenges related to test fragmentation, test-maintenance effort, score calibration, and the risk that an aggregated score may give a misleading impression of system health.

The goal is to design and implement a proof-of-concept for continuous confidence scoring of selected userfacing software features.

The prototype will:

• Identify representative workflows supported by the system.
• Divide each workflow into independently testable user actions.
• Execute automated tests through appropriate application or service interfaces.
• Record test outcomes, execution history, and detected failures.
• Calculate confidence scores for individual actions.
• Aggregate action scores into sub-feature and feature-level scores.
• Reduce scores over a defined period when tests are not executed.
• Report the results through logs, summaries, or a simple dashboard.
• Provide sufficient history to evaluate score changes over time.

The prototype will be evaluated according to:

1) Defect detection capability: whether the tests identify introduced regressions.

2) Confidence-score reliability: whether the scores correspond meaningfully to the actual health of the tested workflows.

3) Testing efficiency: whether the approach can reduce the time required for manual acceptance testing.

4) Maintenance effort: whether the number and granularity of tests remain manageable as the system evolves.

For this thesis proposal, we target students with a strong interest in software engineering, automated testing, and software quality assurance. You are most likely studying a Bachelor’s or Master’s program in Computer Science, Software Engineering, or a related field.

Joakim Trang, Engineering Manager

joakim.trang@axis.com

+46703916483

Certain roles at Axis require background checks, which means applicable verifications will be done in these recruitments. Notice will be provided before we take any action.

We enable a smarter and safer world by creating innovative solutions for improving security and business performance. As a network technology company and industry leader, we offer solutions in video surveillance, access control, intercom, and audio systems, enhanced by intelligent analytics applications.

With around 6,000 committed employees in over 50 countries, we collaborate with partners worldwide. Together, we thrive in our friendly, open, and collaborative culture and inspire each other to think beyond the expected. United by our commitment to inclusion, diversity, and sustainability, we consistently seek to develop our skills and way of working.

For more information about Axis, please visit our website www.axis.com.

Listen to Axis Communications Playlist