Skill UI

This guide details best practices for achieving high standards of reproducibility in visual analytics papers, focusing on open materials, verifiable code, and rigorous documentation. It covers mapping figures to data sources, preregistering studies, documenting data provenance, and defining the level of reproducibility achieved (e.g., turnkey, scripted).

This comprehensive guide outlines the rigorous standards required for achieving experimental reproducibility in systems research papers (e.g., VLDB). It instructs authors on disclosing critical context, including hardware specifications, software configurations, data provenance, workload patterns, and handling measurement variance. Adhering to these guidelines ensures that all performance claims are fully traceable from raw data to published figures, meeting the highest standards of academic review.

A comprehensive guide for authors submitting papers to web conferences. It details how to structure empirical claims, ensuring evidence matches the claim's scope (scale, freshness, generalizability). The guide covers avoiding methodological flaws like temporal or popularity leakage, selecting datasets with proper provenance, and performing rigorous ablations and statistical analysis for maximum academic rigor.

A comprehensive guide for preparing replication and transparency packages for World Development manuscripts. It provides checklists and best practices for both quantitative (code reproducibility, data provenance, de-identification) and qualitative (evidence-to-claim trail, methods appendix) research. It addresses ethical constraints and the unique challenges of double-anonymized review.

A comprehensive framework for ensuring reproducibility in Web Search and Data Mining (WSDM) research. It addresses unique challenges associated with behavioral data, covering data provenance, temporal splitting discipline, bias assumption reporting, statistical rigor (seeds, variance), and the 'honesty ladder' for non-reproducible results, helping researchers publish robust and verifiable work.

Generates a signed, reproducible, and non-PHI audit trail for de-identification processes. This skill provides compliance evidence, including offsets, content hashes, and model provenance, enabling users to verify that sensitive identifiers were correctly redacted for HIPAA/GDPR audits without exposing any plaintext PHI. It is crucial for defensible data handling.