Scientific literature holds a wealth of information and data from decades of research across every field. Often, the literature is far too extensive to allow for access to all the data using a manual literature search alone. For example, there are over 50,000 publications with molecular dyes as one of their topics and the data contained within these publications are expansive. Because of this, the time and expertise that it takes to seek out and find the relevant literature and extract the data is a significant barrier to practically using the data. In this work, we present progress toward the development of a “one-click” tool for web scraping and natural language processing (NLP) of the scientific literature. Using a keyword, web scraping identifies and downloads relevant publications from all the major chemistry publishers (American Chemical Society, Royal Society of Chemistry, Springer, Elsevier, and Wiley). These publications are then subjected to natural language processing using ChemDataExtractor where targeted data is extracted, validated, and stored for later analysis. This tool increases the efficiency with which data can be extracted from the literature. The cyberinfrastructure developed will help to overcome some of the hurdles that stand in the way of a full access to the literature. The large database of chemical structures and properties that ensues can be the subject of new understanding as well as provide new starting points for synthesis of functional molecules and materials.

Toward a “one-click” tool for automated data extraction from the literature

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.