Within-Project Defect Prediction for Ansible Infrastructure-as-Code Scripts

86 open-source GitHub repositories provide metrics for predicting defects in Ansible Infrastructure-as-Code scripts. The dataset includes IaC-oriented, delta, and process metrics, curated from repositories meeting criteria like recent activity and continuous integration. It aims to identify source code and development process properties that are good predictors of defects in IaC.

Use Cases

• Train defect prediction models based on IaC-oriented structural metrics.
• Analyze the impact of code changes on defect likelihood using delta metrics.
• Investigate correlations between development process metrics and defect introduction.
• Benchmark machine learning approaches for software quality assurance in Infrastructure-as-Code.

Strengths

• Data is sourced from 86 real-world, active open-source GitHub repositories.
• Repositories were filtered using 10 specific criteria, including recent activity and contributor count, which likely improves data relevance.
• Metrics are grouped into three distinct categories: IaC-oriented, delta, and process.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

OpenML platform, sourced from open-source GitHub repositories.

Collection Method

Metrics extracted from repositories meeting specific activity, size, and quality criteria.