Web-based tools are developed to help teaching and learning difficult topics in chemistry more interactively. New approaches are developed in teaching and learning organic reactions by making use of artificial intelligence and machine learning. The goal of this project is to stimulate student imaginations to improve learning of difficult concepts in chemistry. Currently, one machine learning tool, the neural machine translation model, is applied to the prediction of organic chemistry reaction products with some success. However, a previously reported application relies solely on a reaction dataset extracted from US Patent Office (USPTO) applications, which causes the model to be unsuitable for the prediction of textbook reactions. For example, the models cannot reliably predict simple substitution or elimination reactions. We present an expanded dataset and trained model based on both the USPTO data and a collection of reactions from several popular introductory organic chemistry textbooks. We have made our datasets and new models publicly available on a GitHub repository. The details of the data collection and model training will be presented.

Expanded reaction dataset to include organic chemistry textbook reactions

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.