ASReview: Software for Active Systematic Reviews

Active Systematic Review (ASReview) is software designed to accelerate the process of systematic reviews. It is written in python, and uses deep learning to predict which papers should be most likely included in the review. Our software is designed to accelerate the step of screening abstracts and titles with a minimum of papers to be read by a human with no or very few false negatives.

ASReview implements an oracle and a simulation mode.

• Oracle The oracle modus is used to perform a systematic review with interaction by the reviewer (the ‘oracle’ in literature on active learning). The software presents papers to the reviewer, whereafter the reviewer classifies them as relevant or not.
• Simulation The simulation modus is used to measure the performance of our software on existing systematic reviews. The software shows how many papers you could have potentially skipped during the systematic review.

The source code is freely available at GitHub.

Basics

• ASReview
• Installation
  • Install Python
  • Install ASReview
  • Server installation
  • Troubleshooting
• Active learning for Systematic Reviews
• ASReview Quick Tour
  • Launching ASReview
  • Create a project
  • Start Reviewing
  • Stop reviewing
• ASReview against COVID-19
  • CORD-19 dataset
  • ASReview plugin
  • Installation and usage
  • License
• Data sets
  • Using your own data
  • Built-in demonstration data sets
• End-user testing
  • Create a project
  • Upload a dataset
  • Prior Inclusions
  • Random papers
  • START reviewing
  • Export data
• Extensions
  • Available extensions
  • How to make an extension
• Frequently Asked Questions
  • Is it possible to get the inclusion likelihood for unlabelled papers?

Advanced

• Command Line Interface (CLI)
  • Simulation mode
• Models
  • nb
  • svm
  • rf
  • logistic
  • nn-2-layer
  • lstm-base
  • lstm-pool
• Query Strategies
  • random
  • uncertainty
  • max
  • cluster
  • mixed
• Balance Strategies
  • simple
  • undersample
  • triple
  • double
• Feature Extraction
  • tfidf
  • doc2vec
  • embedding-idf
  • embedding-lstm
  • sbert

Development

• Extensions
  • Program your extension
  • Create the entry point class
  • Entry points
  • [Optional] Share your extension!
• Application Programming Interface (API)
  • Oracle mode
  • Simulation mode
• Reference
  • Low level API
  • Models
  • Query strategies
  • Balance Strategies
  • Feature Extraction
  • Data
  • Utils
  • State
  • Analysis
  • Extensions

Indices and tables

• Index
• Module Index
• Search Page

Citation

A research paper is upcoming for this project. In the mean time, it can be cited with (fill in x and y for the version number):

ASReview Core Development Team (2019). ASReview: Software for automated systematic reviews [version 0.x.y]. Utrecht University, Utrecht, The Netherlands. Available at https://github.com/asreview/asreview.

@Manual{
        title = {ASReview: Software for automated systematic reviews},
        author = {{ASReview Core Development Team}},
        organization = {Utrecht University},
        address = {Utrecht, The Netherlands},
        year = 2019,
        url = {https://pypi.org/project/asreview/}
}