ASReview has support for extensions, which are a nice way to extend the functionality of the ASReview LAB software or the command line interface. There are officially supported extensions and community contributions.

Officially Supported Extensions

The following extensions are officially supported and were developed as part of the core project:

Model extensions

  • Feature Extraction
    • ASReview-vocab-extractor: This extension adds two feature extractors that extract vocabulary and vector matrices during simulation phases. Might one day be integrated to the core.

Subcommand extensions

  • Visualization
  • Wordcloud
    • ASReview-wordcloud: Creates a visual impression of the contents of datasets via a wordcloud.

  • Statistics
    • ASReview-statistics: Tool to give some basic properties of a dataset, such as number of papers, number of inclusions.

  • Hyperparameter Optimization

Dataset extensions

  • COVID-19
    • ASReview against COVID-19: Makes the literature on COVID-19 directly available in ASReviews so reviewers can start reviewing the latest scientific literature on COVID-19 as soon as possible.

Community-Maintained Extensions

ASReview has support for community-maintained extensions, that enable you to seamlessly integrate your code with the ASReview framework. These extensions can extend the software with new models, subcommands, and datasets.

The following extensions are developed and maintained by the ASReview community:

  • ASReview 17 layer CNN classifier
  • ASReview Model Switcher
    • This extension adds a model that switches between two models during simulation runtime. It can be useful for when later stages of data classification require different models.

    • Github

    • DOI 10.5281/zenodo.5084863

  • ASReview NB + CNN classifier with HPO
    • This extension adds a model consisting out of two separate classifiers for use during simulation mode. The first X amount of iterations (default = 500) are run with a Naïve Bayes model. After the switching, a switch to a CNN is made. Immediately at this switching point, and then after each 150 iterations, hyperparameter optimization is conducted to find the most suitable CNN architecture for current iteration.

    • Github

    • DOI 10.5281/zenodo.5482149

  • ASReview Wide Doc2Vec
    • This small plugin adds a new feature extractor based on doc2vec with a wider vector. In combination with a convolutional neural network model, that has been shown to outperform classical algorithms in some situations.

    • Github

    • DOI 10.5281/zenodo.5084877

  • ASReview matrix and vocabulary extractor for TF-IDF and Doc2Vec
    • An extension for ASReview that adds a tf-idf extractor that saves the matrix and the vocabulary to pickle and JSON respectively, and a doc2vec extractor that grabs the entire doc2vec model.

    • Github

If an extension is not on this list, or you made one and you would like it to be added to this list, please initiate an issue on Github.


If an extension is uploaded to PyPI, it can be installed via command line. In this example, the asreview-visualization extension is used. The extension extends ASReview with functionality for creating plots from the ASReview file.

Install the extension with:

pip install asreview-visualization

If the extension is published on Github, installing directly from the repo can be done with:

pip install git@github.com:{USER_NAME}/{REPO_NAME}.github

See Create an Extension for information about developing your own extension.