A Recommender System Approach for Predicting Drug Side Effects

Diego Galeano Galeano, Alberto Paccanaro

Research output: Chapter in Book/Report/Conference proceedingConference contribution

566 Downloads (Pure)

Abstract

The accurate identification of drug side effects represents a major concern for public health. We propose a collaborative filtering model for large-scale prediction of drug side effects. Our approach provides specific recommendations for side effects of medicines. The proposed latent factor model relies solely on the public drug-side effect relationships from safety data. Applied to 1,525 marketed drugs and 2,050 side effects, we achieved an AUPRC (area under the precision-recall curve) of 0.342 in a hold-out test set, with a sensitivity of 0.73 given a specificity of 0.95, providing state-of-the-art performance in side
effect prediction. Here we also show that our method provides good performance on drug-specific Anatomical Therapeutic and Chemical (ATC) category and side effect- specific medical category of disorders. Our findings suggest that latent factor models can be useful for predicting unknown adverse drug events.
Original languageEnglish
Title of host publicationIJCNN 2018: International Joint Conference on Neural Networks
PublisherIEEE Xplore
Pages1-7
Number of pages7
ISBN (Electronic)978-1-5090-6014-6
DOIs
Publication statusPublished - 2018

Keywords

  • drug
  • side effects
  • recommendation systems
  • Adverse drug reactions
  • latent factor models
  • collaborative filtering

Cite this