To content
Department of Business and Economics

AutoChemplete - Published at W4A'23 & available as Open Source Tool

Image of a molecule © Terry Vlisidis (Unsplash)
Image of a molecule

We are excited to announce that the research group of JProf. Mario Nadj will be represented at this year’s Web for All Conference (W4A’23) with an innovative paper entitled “AutoChemplete – Making Chemical Structural Formulas Accessible”.

The paper, now available in the W4A 2023 proceedings, outlines the development of AutoChemplete – an interactive labeling tool – for chemical structural formulas. AutoChemplete uses machine learning to predict the molecule from an image of the structural formula. It then generates accessible representations of the molecule that can be easily understood by visually impaired students. The tool has been released as open source and can be downloaded from GitHub. Developing AutoChemplete is a notable step forward in helping to make STEM subjects more accessible to visually impaired students.