Subgroup discovery (SD) tries to find relationships between different properties of a dataset with respect to one target variable. SD is an AI tool that ‘stands’ between classification and description. It has a target value, as do classification techniques, but, its objective is to describe a dataset, rather than predict new data. A researcher would use subgroup discovery to find simple, general, and interesting patterns with respect to a target variable of interest. The target variable may be binary, categorical, or continuous. Evolutionary fuzzy systems are a union of evolutionary algorithms and fuzzy logic. Evolutionary algorithms are stochastic algorithms that borrow from natural evolution to solve complex optimization problems. This family of algorithms is capable of efficiently searching large, complex datasets. Fuzzy logic is an extension of traditional set theory, able to model and deal with imprecise data. We will demonstrate how a researcher might use evolutionary fuzzy rule induction processes for subgroup discovery in chemical datasets. Use cases will be presented for identifying structure alerts, describing compound activity, and describing compound selectivity. The input required is a representation for each molecule in the dataset (e.g., SMILES string), and one, or more, target variables. The output is a set of rules, presented both as text and graphically, and quality measures for the rules (e.g., true positive rate, false positive rate). The work will be presented as a collection of R and Jupyter notebooks. All datasets and notebooks will be made available.

Evolutionary fuzzy rule induction processes for subgroup discovery in chemical datasets

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.