Simulation Study Switching between Models
This systematic review focused on synthesizing information on studies that evaluated the performance of Active Learning compared to human reading.
The ASReview Research Team continuously explores and investigates new ways to improve ASReview through simulation studies, use-cases, and more. The team doesn’t do this all alone, but often together with other organizations, research groups, and developers. The findings are published in peer-reviewed journals like Nature Machine Intelligence. By publishing data, scripts and underlying code, all work becomes fully transparent too.
The ASReview project was initiated at Utrecht University by Prof. dr Rens van de Schoot, dr. Daniel Oberski and Prof. dr. Lars Tummers in 2018. The award winning project has now grown into a full multidisciplinary research team that works hand in hand with the ASReview community.
The team works according to the Open Science principles and invests in an inclusive community contributing to the project. In short, research is conducted according to the following fundamental principles:
Utrecht University has established specific regulations governing conduct for its employees. These are based on the key principles of professional and quality academic conduct and ethically-responsible research. Members of the team employed by Utrecht University, commit themselves to these regulations in all their conduct, including all work related to ASReview. Adherence to similar key principles is expected of all researchers involved in all facets of the ASReview project.
This systematic review focused on synthesizing information on studies that evaluated the performance of Active Learning compared to human reading.
ASReview conducted a simulation study on risk analysis documents to evaluate the time-benefit for the Royal Dutch Pharmacists Association.
The MegaMeta project is a large scale project to review factors that contribute to substance use, anxiety and depressive disorders. Read more information on the search and screening protocol, hyperparameter tuning and post-processing used in this post.
This systematic review focused on synthesizing information on studies that evaluated the performance of Active Learning compared to human reading.
The ASReview research team conducted a systematic review on the implementation of AI-aided Systematic Reviews within Clinical Guideline Development.
In a time of exponential growth of new evidence supporting clinical decision making, combined with a labor-intensive process of selecting this evidence, there is a need for methods to speed up current processes in order to keep medical guidelines up-to-date.
This dataset contains an overview of 117 systematic reviews published by corresponding authors affiliated to Utrecht University (UU) or UMC Utrecht in 2020.
ASReview-wordcloud is a supplemental package for ASReview. Wordclouds can help you to get a visual impression of the contents of datasets.
We show that by using active learning, ASReview can lead to far more efficient reviewing than manual reviewing, while exhibiting adequate quality. Furthermore, the presented software is fully transparent and open source.
Explore the systematic review dataset that was used for the publication “Psychological theories of depressive relapse and recurrence” from Brouwer et al., 2019. From pre-processing to the final dataset, a look into the complete systematic review process behind this publication.
Recreate the simulation study on the systematic review of Smid et al.. From pre-processing to the final dataset, dive into the complete process behind this publication.
After manually screening 5050 studies, Smid et al included only 27 studies for their review. In the current study, both the Bayesian and logistic regression models found more than 80% of relevant publications after screening only 10% of all publications.
To determine the defaults we performed a simulation study and Naive Bayes + TF-IDF model performed the best.
Combining human intelligence and machine learning into Researcher-in-the-loop machine learning. An effective technique for training models.
The CORD-19 database is available in ASReview and can be used to search for relevant Corona-related publication using active learning.
ASReview LAB is user-friendly software for exploring the future of AI in systematic reviews. The software implements an Oracle Mode, an Exploration Mode, and a Simulation Mode.
All the documentation surrounding ASReview, from API to the Zen of ELAS, can be found within the Read the Docs. If you want to know more about ASReview, this is the place to be! Main topics include: An introduction, ASReview LAB, Features, Extensions & API