Applied data science and machine learning (ML) are making differences in the chemistry domain including molecular design of drugs and materials as well as prediction and optimization of synthesis routes and production processes. To train effective ML models for these emerging areas, a large amount and a variety of well-curated experimental datasets are essential. Among such datasets needed in chemistry, the structured and curated datasets are oftentimes locked down in proprietary indexing databases, corporate in-house databases, or reference works, either in print or digitized; while the unstructured data scattered in the published research articles. In this presentation, we will first analyze the recent publications reporting ML research in chemistry and examine some ML research projects at MIT to highlight the types and extent of chemistry datasets needed for ML. Then, we will discuss the availability of these datasets and texts as well as the economic, legal and technical challenges in obtaining access to these datasets for data and text mining purposes. To tackle these challenges, we will share some strategies on exploring and developing collaborations among researchers, librarians, publishers, and database providers.

Chemists’ data needs for machine learning research in academia

Macromolecular Chemistry: The Second Century

[Author Benefits](https://scimeetings.acs.org/?utm_source=pubs_content_marketing&utm_medium=website&utm_campaign=0320_MCF_NPI_Launch_Spring_Homepage&ref=pubs_content_marketing)

[How to use SciMeetings](https://storage.googleapis.com/dl.morressier.com/acs/SciMeetings-ACSSpring2020-HowToGuide_M3020-Final.pdf)

[Frequently Asked Questions](https://storage.googleapis.com/dl.morressier.com/acs/SciMtng_ACSNMSp2020_FAQ_M3120_Final.pdf)

ACS Spring 2020 National Meeting & Expo

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