Coordinating Projects using ASReview

LocationUtrecht
Duration84 hours (of which 56 hours of self study)
Starting momentStart date will be determined through mutual agreement
LanguageEnglish
Teaching methodOnline, At location
CertificationMicrocredentials, Certificate, Accreditation points
Price€2,250
ECTS3. EC

This expert-led course will provide you with the necessary knowledge and skills to supervise AI-aided screening projects within your organization. The course includes a globally recognized EduBadge upon completion. Invest in your career and stay ahead in your field.

Course contents

The scientific output of the world doubles every nine years. In this tsunami of new knowledge, imagine the enormous challenge of updating a medical guideline, developing evidence-based treatments, creating unbiased policy recommendations, or scouting for new technologies. There’s not enough time to read everything! Attempts to synthesize state-of-the-art research findings in a particular scientific field therefore are under severe strain; the exponential growth in papers means screening an increasingly larger body of work – resulting in costly, or error-prone work that is often abandoned prematurely. Nevertheless, systematic reviews are pivotal for scholars, clinicians, policymakers, journalists, and, ultimately, the general public.

The rapidly evolving field of Artificial Intelligence (AI) has allowed the development of AI-aided pipelines that assist in finding relevant texts for such search tasks. A well-established approach to increase the efficiency of text screening is screening prioritization via active learning: an interaction between a human screener and a machine learning model which constantly learns from the human and proposes the most likely relevant paper to the human. This method can reduce the number of papers to screen by up to 95%(!) and is easily integrated in popular, active learning text screening software.

Our course prepares the participants for coördinatingprojects which are carried out in the open-source software ASReview.

Target audience

This course is designed for project coordinators, data scientists, information specialists, PhD supervisors and PIs of bigger projects, and others who wish to coördinate AI-aided systematic reviews.

Prior experience with traditional systematic reviews is expected and we assume basic knowlegde of ASReview obtained via, for example, the summer school course or self-study.

Course goal and learning goals

The primary goal of this course is to equip participants with the knowledge and skills required to coordinate AI-aided screening projects using ASReview within their organizations. The course accommodates individual learning objectives and projects, ensuring that upon completion, participants will be able to:

Comprehend the basics of the active learning cycle and the human’s role within the loop;
Understand the components of the active learning model, and learn how to select appropriate training data;
Illustrate how AI-aided screening integrates into the existing pipeline within a particular organization;
Apply the FAIR data principles to the outcomes of an AI-aided systematic review;
Collaborate with the community by participating in related research projects, contribute to the software ecosystem, write about new applications, or generate knowledge in other ways;
Effectively communicate benefits (and pitfalls) of AI-aided systematic reviews to stakeholders.

Costs

€ 2,250 (standard tariff)

Active participation and attendance

We expect active participation and attendance in at least 80% of the sessions. A repair option is available for any incomplete or missed assignments, to ensure participants can meet the certificate requirements.

View the detailed programme below.

PART 1: Theory (22 hours)

Meeting 1 (8 hours)

Meet & greet the other participants + explain the set-up of the course + lecture on the steps of systematic reviewing and where AI-aided screening fits in + differences between software packages + intro on active learning

Optional: Meet the core developers and information experts and join the discussions during the weekly stand-up

Self-study (6 hours)

Read literature + watch instruction videos + exercise in using ASReview LAB

Optional: Try out ASReview LAB with your own dataset