Application of AI Technologies for Promoting the Heritage of Outstanding Scientists by Scientific Libraries: on the Example of A. E. Favorsky’ Works

Abstract

The purpose of this article is to present a new method for preserving and promoting the scientific heritage of chemists: creation of an intelligent system (IS), based on artificial intellect (AI), for processing chemical literature, on the example of publications by Academician A. E. Favorsky. Libraries employ various models to preserve scientific heritage and popularize science – ranging from lectures, exhibitions, memorial libraries of scientists, and bibliographic databases of research publications to modern formats based on state-of-the-art generative AI patterns. The authors briefly describe the experience of the Central Scientific Library of the Federal Research Center Irkutsk Institute of Chemistry (Siberian Branch of the Russian Academy of Sciences) named after A. E. Favorsky in preserving the scientific heritage of chemists. The development of AI systems in the field of chemistry requires the creation of independent datasets based on verified information, as well as the development of intelligent systems for the machine analysis of chemical literature. The article justifies the selection of the works of an eminent organic chemist – Academician A. E. Favorsky – as a reliable primary source for creating benchmark-training datasets intended for the digitization and subsequent recognition of chemical graphs within his texts. It presents AI tools, being developed by the library for chemical text recognition, specifically the application of Optical Chemical Structure Recognition (OCSR) technologies to convert molecular graphs into a machine-readable format (SMILES). The library’s work in creating modern intellectual tools not only brings historical knowledge into digital circulation but also sparks the interest of younger generations in the profession of a research chemist.