Sysrev

Sysrev facilitates FAIR-compliant data curation and systematic evidence review by combining human review with machine learning to extract structured data from digital documents.

Key Features:

• Systematic Evidence Review Support: Provides a structured workflow for conducting transparent and reproducible systematic evidence reviews on digital documents.
• Data Curation Projects (Sysrevs): Enables creation of "sysrevs" where users upload documents, define review tasks, and automate components of the review process.
• Human–Machine Learning Integration: Leverages human expertise together with machine learning algorithms for generalized data extraction from unstructured or siloed datasets.
• Automation and Redundancy Reduction: Automates specified review processes and reduces redundant human effort in evidence synthesis workflows.

Scientific Applications:

• Systematic evidence synthesis: Supports systematic evidence reviews and evaluation of digital documents to produce reproducible review outcomes.
• FAIR data curation: Facilitates creation of FAIR-compliant datasets to improve findability, interoperability, and reuse of research data.
• Cross-disciplinary data-intensive research: Manages large volumes of digital documents and datasets applicable to healthcare, environmental science, and social sciences.
• Extraction from unstructured data: Enables extraction of structured insights from unstructured or siloed information using combined human and machine approaches.

Methodology:

Combines human review with machine learning algorithms for generalized data extraction and supports user-defined review tasks with automated processing of those tasks.