asreview 2.2

ASReview LAB - A tool for AI-assisted systematic reviews

🎉 ASReview LAB v2 is here! 🎉
Faster, smarter, and more flexible than ever before.
Discover the new AI models, improved workflow, and enhanced user experience.

ASReview LAB: Active Learning for Systematic Reviews

ASReview LAB is an open-source machine learning tool for efficient, transparent, and interactive screening of large textual datasets. It is widely used for systematic reviews, meta-analyses, and any scenario requiring systematic text screening.

The key features of ASReview LAB are:

• Active Learning: Interactively prioritize records using AI models that learn from your labeling decisions.
• Scientifically validated: ASReview LAB has been scientifically validated and published in Nature Machine Intelligence.
• Flexible AI Models: Choose from pre-configured ELAS models or build your own with custom components.
• Simulation toolkit: Assess model performance on fully labeled datasets.
• Label Management: All decisions are saved automatically; easily change labels at any time.
• User-Centric Design: Humans are the oracle; the interface is transparent and customizable.
• Privacy First: Everything is open source and no usage or user data is collected.

---

What's New in Version 2?

On May 14th, ASReview LAB version 2 was released with a large set of new features. The most notable new features are:

• New ELAS AI Models: Pre-configured, high-performance (+24%) models for different use cases (Ultra, Multilingual, Heavy). More new and exciting models can now be found in our new ASReview Dory extension.
• Improved User Experience: The interface is faster, progress monitoring is better, and there are more customization options (such as dark mode, font size, and keyboard shortcuts).
• ASReview LAB Server with crowd screening: Screen a single project with multiple experts. All users interact with the same AI model.
• Quick project setup: Start screening new datasets in seconds using the quick setup for projects.
• Add customizable tags: Add tags and groups of tags to your records and label decisions. This makes data extraction much easier!
• Improved simulation API: The new and flexible simulation API opens up a whole new simulation potential. It is a perfect tool for hunting for even better-performing models.

---

Installation

Requires Python 3.10 or later.

pip install asreview

Upgrade:

pip install --upgrade asreview

For Docker and advanced installation, see the installation guide.

Latest version of ASReview LAB:

The ASReview LAB Workflow

1. Import Data: Load your dataset (CSV, RIS, XLSX, etc.).
2. Create Project: Set up a new review or simulation project.
3. Select Prior Knowledge: Optionally provide records you already know are relevant or not relevant.
4. Start Screening: Label records as Relevant or Not Relevant; the AI model continuously improves.
5. Monitor Progress: Use the dashboard to track your progress and decide when to stop.
6. Export Results: Download your labeled dataset or project file.

---

Documentation & Resources

• Documentation
• Video tutorials
• AI models of ASReview LAB
• FAQ
• Live Demo

Citation

If you wish to cite the underlying methodology of the ASReview software, please use the following publication in Nature Machine Intelligence:

van de Schoot, R., de Bruin, J., Schram, R. et al. An open source machine learning framework for efficient and transparent systematic reviews. Nat Mach Intell 3, 125–133 (2021). https://doi.org/10.1038/s42256-020-00287-7

For citing the software, please refer to the specific release of the ASReview software on Zenodo: https://doi.org/10.5281/zenodo.3345592. The menu on the right can be used to find the citation format you need.

For more scientific publications on the ASReview software, go to asreview.ai/papers.

Community & Contact

The best resources to find an answer to your question or ways to get in contact with the team are:

• Newsletter
• FAQ
• Community events
• Issues or feature requests
• Donate to ASReview
• Contact ([e-mail])

License

The ASReview software has an Apache 2.0 LICENSE. The ASReview team accepts no responsibility or liability for the use of the ASReview tool or any direct or indirect damages arising out of the application of the tool.