Autumn School "From Data to Graphs: Workshop on Bibliometric Network Analysis"

How to avoid drowning in an ocean of scientific literature?
Characteristics of modern science—the increasing complexity of knowledge, the growing specialization and fragmentation of scientific disciplines, and the growing number of researchers and research groups worldwide—make the task of searching for literature far from trivial. Science is advancing rapidly, and the number of new studies is growing exponentially. The nature of modern research is becoming increasingly complex and combines largely opposing trends: on the one hand, increasing specialization within individual scientific fields, and on the other, growing interdisciplinary interaction and the emergence of new scientific fields. Beyond these global trends, there is a diversity of formats and sources for discovering scientific knowledge. Scientific papers are published not only in the form of books and peer-reviewed journal publications, but also as conference proceedings, preprints, reports, dissertations, and other materials. Bibliographic information is contained in scientific citation databases, and today there is a wide variety of such databases – from the largest and most authoritative (Google Scholar, Web of Science, Scopus) to relatively new and recently developed ones (OpenAlex, Lens.org, Dimensions, ResearchGate). Databases vary in the extent of their discipline coverage and the types of materials they include, which affects the completeness and representativeness of the scientific information they contain. This also complicates searching and systematizing information, as there is a risk of missing important works. Finally, researchers' resources are also limited – scientists around the world have long since become accustomed to the impossibility of reading and deeply analyzing all published works in their field.

How can automated tools, including AI, be effectively used in searching for scientific publications?
The development of large language models (LLMs) and natural language processing (NLP) technologies has given a significant boost to the implementation of artificial intelligence (AI) tools in the literature search process. This has significantly transformed the way students and researchers work with scientific literature in a rapidly changing academic landscape. New AI-based systems such as Connected Papers, Semantic Scholar, Elicit, Consensus, ChatPDF, Iris.ai, and Research Rabbit allow not only searching but also summarizing published works. Any new technology carries both risks and opportunities, so it's always a good idea for any researcher to understand the advantages of such systems.  

At the Autumn School, participants will learn about methods for constructing and interpreting bibliometric networks and the capabilities of modern scientific databases and systems based on AI and natural language processing. Participants will also learn how to use bibliometric network analysis to support their own research projects. The classes will be taught by specialists with recognized expertise in bibliometrics and network analysis. Participants will engage in practical assignments, work with real data, and discuss research cases.  

Registration is open until November 25th. Space is limited, so apply now!

The school is part of the project "From Search to Analysis: A Complete Guide to Working with Scientific Literature," which is being implemented by a team led by Anna Mikhailovna Semenova, a junior research fellow at the International Center for Decision Analysis and Choice. The project team includes Irina Anatolyevna Pavlova and Natalia Nikolaevna Matveeva, both employees of the International Laboratory of Decision Analysis, and Daria Vasilyevna Maltseva, who is as an expert.