7. Cluster Analysis of Drugs and their Adverse Effects

Brittney Man Butler University
Faculty Sponsor(s): Rasitha Jayasekare Butler University
Cluster analysis is an unsupervised technique that groups observations together based on similarities and dissimilarities to one another. Clustering can lead to uncovering meaningful patterns among the data. The goal of this project is to cluster drug information, patient reaction, and patient outcome data from the FDA’s Adverse Events Reporting System (FAERS). Utilizing the measures such as Gap Statistics and Silhouette Coefficient, the optimal number of clusters and how well the data are clustered will be determined. The clusters created from statistical methods such as K-Means and Agglomerative Hierarchical Clustering will then be compared and analyzed for nontrivial patterns or specific variables that may drive the clustering to better understand the relationships among the FAERS data.
Mathematics & Computer Science
Poster Presentation

When & Where

Irwin Library 3rd Floor