In this talk, we will detail some of the work at IBM Research - Almaden focused on applying tools of cognitive computing and machine learning to chemistry and materials science. Modern machine learning algorithms such as deep learning are hungry for data -- often requiring datasets much larger than those currently available in many areas of materials science. The areas of research in our group include information extraction for materials science data, knowledge representation and exploration in materials science datasets, multi-category labeling problems in materials data, and integration of human expert knowledge for materials discovery. The main focus area of the research is soft materials, though there are active collaborations in other areas, such as inorganic functional materials.

Cognitive computing and machine learning for chemical & materials discovery

255th National Meeting (2018)

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