AI System Proposes and Evaluates Molecular Structures from Spectroscopic Data
A new artificial intelligence (AI) system has been developed to propose suitable molecular structures directly from raw spectroscopic measurements. The system is also capable of assessing the plausibility of these proposed structures. It is openly accessible to researchers and has been published in the journal Nature Communications by a collaborative team of institutions.
A research team has developed an innovative artificial intelligence (AI) system designed to process raw data from spectroscopic measurements. This system's primary function is to propose suitable molecular structures based on the input data.
Beyond generating potential structures, the AI system further evaluates their plausibility, contributing to a more efficient and accurate analysis process. The developers have made this system openly accessible to the scientific community.
The findings and details of the AI system's development have been published in the journal Nature Communications. The collaborative effort involved researchers from Friedrich Schiller University Jena, Helmholtz-Zentrum Berlin for Materials and Energy, and the Helmholtz Institute for Polymers in Energy Applications Jena. Additionally, the Swiss software company Zakodium Sárl contributed to the project.
According to Phys.org, this development aims to streamline the evaluation of chemical spectra.
